Effects of Synthesized Voice Warning Parameters on Perceived Urgency

Abstract

The effects o f synthesized voice w a rn in g param eters on perceived urgency were exam ined in order to build a detailed and usable description of the relation between the param eters of synthesized voice w arnings and perceived urgency. Ten native and 10 non-native English speakers participated in 4 experiments to evaluate and q u a n tify the effects o f the voice parameters. The results showed that speech rate, average fundam ental frequency (F0), voice type, and fundam ental frequency co n to u r have clear effects on the perceived urgency of synthesized voice w arnings. The effects of quantitative param eters on perceived urgency were scaled using an a p p lica tio n of Stevens's power law (1957). In addition, the results showed sig n ifica n t differences in the perceived urgency o f average F0 and F0 co n to u r types between native and non-native

Designing acoustic and non-acoustic parameters of synthesized speech warnings to control perceived urgency

“…Magnitude estimation is the psychophysical ratio scaling most frequently used in current investigations and has been used to describe the relationship between sensation magnitude and stimulus magnitude, such as for brightness (Stevens and Guirao, 1963), loudness (Stevens, 1957) and even for passenger preferences of train interior designs (Han et al, 1998). Recent researches have demonstrated that applying this method is a viable experimental technique to quantify the relations between perceived urgency and non-verbal auditory warning parameters (Hellier et al, 1993), and between perceived urgency and synthesized speech warnings generated by simple rulebased text-to-speech (TTS) system (Park and Jang, 1999).…”

Analyzing crowdsourced ratings of speech-based take-over requests for automated driving

“…No clear gender effects seem to exist: words spoken by a female typically yield similar urgency ratings as the same words spoken by a male (e.g., Hellier et al, 2002;Wogalter et al, 2002). However, Jang (2007) and Park and Jang (1999) found that a male voice yielded higher urgency ratings than a female voice. Furthermore, interaction effects have been observed, where the word "Note" received a higher urgency rating when spoken by a male instead of a female (Hellier et al, 2002).…”

“…Second, emotional tone has important effects: phrases are considered more urgent if spoken in an urgently intoned style (Edworthy et al, 2003a;. Third, it has been found that the greater the speech rate, the higher the perceived urgency (Hollander and Wogalter, 2000;Jang, 2007;Park and Jang, 1999). No clear gender effects seem to exist: words spoken by a female typically yield similar urgency ratings as the same words spoken by a male (e.g., Hellier et al, 2002;Wogalter et al, 2002).…”

Toward Online Virtual Safety Training in the Manufacturing and Service Industries

“…Using the free-modulus technique (Hellier, et al, 1995;Park and Jang, 1999) subjects were asked to give their subjective rating of hazard, acceptability and percent chance of tool breakage when a part is machined on a DynaMyte 2400 High Precision Computer Numerically Controlled (CNC) milling machine.…”